Our Predictions for 2024 AI transcript and summary - episode of podcast After Hours

Go to PodExtra AI's episode page (Our Predictions for 2024) to play and view complete AI-processed content: summary, mindmap, topics, takeaways, transcript, keywords and highlights.

Go to PodExtra AI's podcast page (After Hours) to view the AI-processed content of all episodes of this podcast.

After Hours episodes list: view full AI transcripts and summaries of this podcast on the blog

Episode: Our Predictions for 2024

Author: TED Audio Collective / Youngme Moon, Mihir Desai, & Felix Oberholzer-Gee
Duration: 00:49:25

Episode Shownotes

What’s in store for 2024? Mihir and Felix are back with their celebrated predictions episode. Will OPEC implode? Are quant funds in trouble? What’s Argentina’s future? Can inflation in the U.S. really sink to 2%? Is plastic the new asbestos? Who will acquire Electronic Arts? Is AIML a verb? Listen

in as the hosts (foolishly) predict what the new year will bring. Hosted on Acast. See acast.com/privacy for more information.

Full Transcript

00:00:00 Speaker_01
If you're looking for flexible workouts, Peloton's got you covered. Summer runs or playoff season meditations. Whatever your vibe, Peloton has thousands of classes built to push you. We know how life goes. New father, new routines, new locations.

00:00:16 Speaker_01
What matters is that you have something there to adapt with you whether you need a challenge or rest. And Peloton has everything you need, whenever you need it. Find your push. Find your power. Peloton. Visit OnePeloton.com

00:00:44 Speaker_00
Hello, everyone. You're listening to After Hours. I'm Felix. I'm Mihir. And it's time for predictions, Felix. It is time for predictions. You know how you predict all kinds of things all year long, really? Not only at the end of the year.

00:00:58 Speaker_00
Do you ever go back? To check, are you mostly right, mostly wrong?

00:01:04 Speaker_03
You know, I don't that often. I think part of the fun of it is just thinking about the future. I don't think you necessarily really want to be constrained by the mistakes you've made in the past.

00:01:14 Speaker_03
Having said that, obviously, as you point out, every day we make a thousand predictions. And naturally create intuition out of all those little predictions. And I think that's a very powerful process.

00:01:26 Speaker_03
And so I don't think I've ever formally gone back and really revisited it, but I think predictions are a way of honing intuition, which has got to be one of the most important skills in life. That's in part what makes it fun.

00:01:39 Speaker_00
Yeah, I'm exactly like you. I rarely go back. That's always been really fascinating to me, how much people are influenced by what actually happens. So exposed, you see the company got bought or the company didn't get bought.

00:01:55 Speaker_00
But really, what it's mostly about is, was your thinking right? If the thinking was correct, that led you to make a prediction. Because it's all uncertain. It's all probabilistic anyway. And so you shouldn't be swayed very much. Your thinking was right.

00:02:09 Speaker_00
It was 50-50. And I guess half the time, what happens in the end is not what you predicted. But you shouldn't really be that influenced.

00:02:18 Speaker_03
It's tempting sometimes to look at ex post outcomes and be like, oh, I was wrong. And it's not always the right way to think about the problem. It's not right. Yes. Although it's so tempting and people tend to do that. Yes.

00:02:29 Speaker_03
Well, I think this gives us license to go crazy, Felix.

00:02:32 Speaker_00
Yes. Let's do it.

00:02:39 Speaker_03
Alright, Felix, your first prediction.

00:02:41 Speaker_00
Yes, so my first prediction is in 2024, we will see OPEC, the oil cartel, we will see it fall apart. And here is how I'm thinking about it.

00:02:56 Speaker_00
So if you're in a cartel and you can raise prices by cutting production, that's of course super interesting and lucrative if there is a big price increase as a result of the more limited availability of oil.

00:03:12 Speaker_00
As we diversify the sources of energy, so there's more electricity, there's more solar, there's more wind, there's more everything, I think that volume reduction will be much, much bigger.

00:03:26 Speaker_00
And as a result, it will be much more difficult to keep the cartel intact. Am I super sure that this is going to happen in 2024? No. But directionally, I think we will see OPEC being a much, much weaker force in the global economy.

00:03:45 Speaker_03
Interesting, Felix. So that depends on two things. It seems like one is you seem to be saying that external dynamics will not favor them, meaning as renewables grow, their shares come down and their power goes down.

00:03:59 Speaker_03
Do you have a vision also of what's happening internally? So you're not telling a story of they start to fight and it falls apart. You're telling a story of the external environment is not conducive to their power.

00:04:10 Speaker_00
Yes, that's mostly how I'm thinking about it. Because if demand is very price inelastic, then agreeing on production cuts is relatively easy. The moment the demand is much more elastic, now you will argue

00:04:26 Speaker_00
as we know from the past, it's not easy for them to agree on production cuts in the first place. And I think these tensions will just be exacerbated.

00:04:34 Speaker_00
I have less insight into how the negotiations are going to evolve, but I just think the cartel as a force in the world will be much more limited.

00:04:44 Speaker_03
Wow, fantastic. That's a super interesting one. I might take the other side of that or certainly take it over a longer time. Mainly because I think of the internal dynamics. I think primarily it's internal dynamics that dictate that.

00:04:58 Speaker_03
And so one way to interpret it is that as their share falls, small countries become less important and then the big swing producer becomes more important and then they can enforce discipline.

00:05:11 Speaker_00
But that's a super interesting prediction.

00:05:13 Speaker_00
My sense was it's less about the total share, but it's about the response of the world to an increase in oil prices, where we now have degrees of flexibility that just didn't really exist because we didn't have the battery systems, we didn't have the diversity of sources.

00:05:31 Speaker_00
And so the more we can respond to an increase in oil price, the less attractive it becomes to create that increase in oil prices in the first place.

00:05:40 Speaker_03
Super interesting. I love that prediction.

00:05:41 Speaker_00
Who knows? Definitely one to be potentially wrong.

00:05:46 Speaker_03
What's your first prediction, Mier? So we got to do something on AI, and here comes mine, which is everything in AI will become longer and shorter and narrower and deeper. So first, I think things in AI will get shorter.

00:06:03 Speaker_03
And in particular, I think leads will get shorter. For example, currently, Chat GPT and OpenAI look like they're well ahead of competitors. NVIDIA looks like they're well ahead on computing power. I think all those leads shrink.

00:06:19 Speaker_03
and everything gets shorter. At the same time, I think some things get longer, which is I think the productization and how long people take to adopt takes considerably longer.

00:06:32 Speaker_03
So I think Adobe in their most recent earnings call actually warned a little bit on this. And so we're going to see things get a little bit longer as well. In particular, adoption is just going to take much longer than people think.

00:06:44 Speaker_03
And then I think things get narrower and deeper. And so narrower means we become much more clear about the specific pieces of the economy where AI will pay off and the kind of informational problems that AI will really help us with.

00:07:00 Speaker_03
And then there'll be large chunks of the economy where it'll become more clear that AI will not have an impact. And then the places where it will, it will get really deep and it'll get really profound.

00:07:12 Speaker_03
So I guess I have a sense that in just in general, we have seen AI blanket everything in the world. And I think what we're gonna see is everything gets shorter, leads get shorter. things get longer, narrower, and deeper.

00:07:26 Speaker_03
And then my final prediction is one I'm personally invested in, which is AML, A-I-M-L, becomes a verb.

00:07:34 Speaker_00
And so we'll now talk about AML-ing everything. Oh, wonderful. Yeah, I think I agree with so much of what you said, in particular the part about the differences in the quality of these systems. I think they will just disappear over time.

00:07:49 Speaker_00
It'll actually be really interesting to see competition between OpenAI and Tropic, Google. And I think essentially it will be maybe a little bit like the situation in cloud.

00:08:03 Speaker_00
Companies compete on an almost equal footing and then they carve out niches, things that they're really good at.

00:08:10 Speaker_00
The part that I find really interesting is about where it will and where it will not lead to productivity increases, which in the end dictates where it's going to be big and where it's not going to be big.

00:08:22 Speaker_00
And for many of the super optimistic predictions, I have sometimes a sense that people don't really carefully identify bottlenecks.

00:08:32 Speaker_00
So for instance, take all the excitement about I can write so many more emails in half the time, a third of the time, 20% of the time. Yeah, that's exactly right. But what about the recipient? What about the people who have to read the emails?

00:08:51 Speaker_00
I don't know about your experience, I can already tell pretty much when someone uses chat GPT to write an email, and then because it's costless to them, these emails tend to be really, really long. And sometimes they're really, really complex.

00:09:05 Speaker_00
And so I'm sitting there and instead of reading five lines, I now have a small book ahead of me. And that, of course, will undo many of the productivity gains that we expected in the first place.

00:09:17 Speaker_00
So what I always try to do is think through the entire process and then think about where is that bottleneck? AI is really productivity enhancing if it helps with the bottleneck. But generally speaking, yeah, I think it'll be super interesting.

00:09:33 Speaker_00
Both the width and I guess the depth of the technology that will change over time. That's a really nice prediction.

00:09:39 Speaker_03
Fantastic. Okay. What do you got, Felix?

00:09:42 Speaker_00
We will see a tsunami of erotic and even pornographic content that is produced by artificial intelligence. Right now it's basically held in check. because the big reputable companies are really concerned about that kind of output.

00:10:05 Speaker_00
So, for instance, when the systems first became available, it turned out they were fabulous at writing erotic novels. Companies like Microsoft, companies like Google, Alphabet, they have their reputation at stake and as a result, they cut down.

00:10:22 Speaker_00
Just a couple of weeks ago, Twitch allowed its users to use artistic depictions of nudity. And after a couple of days, they had to adjust the policy because they got flooded with content. But I think there's enormous possibilities.

00:10:40 Speaker_00
There's obviously demand as we know from the web more generally. And as AI, the technology moves from large reputable companies to more niche players who have less of a reputation.

00:10:53 Speaker_00
Maybe it's the traditional pornographic companies that we have today that will fully exploit the technology. We have no idea how much content will be created. That I think will be

00:11:06 Speaker_00
as dominant in AI-generated content as pornography is now dominant on the web itself.

00:11:13 Speaker_03
It's so interesting. It's like one of those dirty little secrets about the web that pornography was so important in its growth and just continues to be so important. I don't think we understand always fully how important pornography is to the web.

00:11:29 Speaker_03
And so I think you're absolutely right. And of course, you're pointing to this larger theme, which is just the cheapening of content creation with AI. And so pornography is going to be one place where we see it. It's going to be with political news.

00:11:41 Speaker_03
It's going to be with potentially fake news. And so that I think is going to lead to maybe more guardrails, but just maybe more of a coarsening of dialogue, and maybe a little bit more of concern and skepticism about the power of these things.

00:11:58 Speaker_03
So I think you're absolutely right. And it's gonna be, in some sense, kind of fairly unpleasant to watch, but what do you make of AI's ability to actually foster more human connections?

00:12:10 Speaker_03
So there are now conversations about friendships with AI bots, that becoming, a more fertile area.

00:12:18 Speaker_00
I think it's pretty amazing already. I guess depending on your personality, it will take on many, many different forms, including sexual and erotic interchanges. To give you one personal example, so character AI, you probably know.

00:12:33 Speaker_00
So I recently used it and when you first open it, it has some famous celebrities, if you don't want to create your own companion. So I happened to see Einstein.

00:12:44 Speaker_00
So I called out Einstein and my prompt was, I think this whole theory about dark matter is complete nonsense. You just made it up. And then a 20 minute conversation about what we know about dark matter and why we think it exists.

00:13:01 Speaker_00
All that was missing from it was the slight German accent in the way Einstein really spoke. But the whole back and forth.

00:13:09 Speaker_00
really just fabulous and I learned a bunch of things about dark matter and he didn't probably learn anything because he already knew everything there is to know.

00:13:20 Speaker_03
That's a great prediction. I love it. That's awesome.

00:13:22 Speaker_00
So what is your next prediction Mihir?

00:13:25 Speaker_03
There is this proverb, the longest mile is the last mile home. And I think that's roughly true. And I think we're going to see that in the economy. So this is a prediction about the macro monetary policy, fiscal policy outlook.

00:13:38 Speaker_03
I think the last mile, of course, is getting from 3% inflation in the US to 2% inflation. And I think that last mile is going to be a long mile and have lots of zigs and zags and actually be more complicated than people understand it to be.

00:13:54 Speaker_03
So if that's right, then we don't get to 2% that fast. And we get there and then we go back, we kind of stay closer to three. So there'll be a big debate about whether three is the right number as opposed to two being the right number.

00:14:06 Speaker_03
I think the Federal Reserve will find itself subject to political pressure in a way that it may never have before. Because during a very complex election year,

00:14:20 Speaker_03
central banks around the world, but in particular, the Federal Reserve will find themselves in the crosshairs of politicians. They will feel even more pressure to become politicized in a way that they have never seen before.

00:14:35 Speaker_03
And then the final piece of that story is We had this remarkable event in October where bond yields blew out up to 4 or 5%. And I think that is not a red herring. I think that is the canary in the coal mine.

00:14:52 Speaker_03
We're going to see those concerns reassert themselves. And I think broadly, what monetary policy was in the last 15 years, which is kind of the dominant thing in the world, is going to be replaced by fiscal policy.

00:15:06 Speaker_03
And the next 15 years are about fiscal policy in the way that monetary policy was dominant for the last 15 years.

00:15:12 Speaker_00
Yeah, interesting. Can I ask you about the first part of your prediction? So say if you go from 4% to 2%, that's really hard. What is it that is substantively different from, say, going from 6% to 4% or 8% to 6%?

00:15:28 Speaker_03
That's a good question. I don't think it's about the difference between six and five and three and two.

00:15:32 Speaker_03
I think it's more that two, which we lived with for so long, is kind of unsupportable going forward, which is that was undergirded by lots of goods deflation, especially from China.

00:15:44 Speaker_03
So I don't think it's about what is this one percentage point difference, but if you have wage growth for the reasons we talked about in the last episode, and you have pressure on that margin, it's just very difficult to get all the way back down.

00:15:57 Speaker_03
So said another way, the fact that we lived for so long at two might just have been an artifact of very odd circumstances. I think we may need to just renormalize around a different number. And that debate will, I think, become very, very political.

00:16:12 Speaker_00
Yeah, and this is a great debate to have because, as you pointed out, there's nothing magical about two.

00:16:18 Speaker_00
In fact, one of the most important reasons why we like low inflation numbers is because unexpected inflation has really big consequences, has really big costs.

00:16:30 Speaker_00
As long as we're in steady state and the number is 2 or 3 or 4, but our expectations match up with reality, I don't really think there's anything that special about a number like 2, it's just that

00:16:45 Speaker_00
As inflation goes up, it typically becomes much more volatile. And as inflation bounces back and forth, then our expectations are not in line with what happens, which always leads to big redistributions that we want to avoid. Exactly.

00:16:59 Speaker_00
You could make many other arguments. So for instance, we know it's very hard to cut nominal wages. If you have a little bit of inflation, there's a lot of flexibility.

00:17:09 Speaker_03
Yeah, I totally agree with everything you said. The contrasting force is political, which is I think inflation has proven to be much more politically costly than anybody anticipated.

00:17:20 Speaker_03
Now, of course, that was very high inflation relative to the levels we're talking about.

00:17:24 Speaker_03
But I think there is going to be this pushback about how costly inflation really is because people are angry about it and they seem to be persistently angry about it. So I think you're right.

00:17:35 Speaker_03
In a normal world, we'd have a conversation about whether two or three makes sense. But I think there is also going to be this pushback, oddly, on the politics of it, which is, gosh, is it more costly than we thought it was?

00:17:45 Speaker_03
This kind of idea that we're returning back to the world that we had before of 2% very quickly, I think is going to turn out to be wrong. And that'll be interesting part of 2024. Super interesting. Okay, Felix, rock on.

00:17:59 Speaker_00
My prediction is that one of the big streaming companies, so Disney, Netflix, will buy Electronic Arts, the electronic game maker. Think about it, what happens in streaming.

00:18:15 Speaker_00
So one discovery is that if you add a tier that is advertising supported, that can be financially very lucrative. And in fact, if you look at revenue per subscriber, it's probably even better than having fairly high subscription prices.

00:18:32 Speaker_00
So I see Disney+, I see Netflix, the basis of their model is an ad-supported tier.

00:18:39 Speaker_00
And then the next trick that we talked about once, because I find it so irritating, is that instead of giving us entire shows and entire seasons, they now drip, drip, drip, give us little pieces. So think about what that means.

00:18:55 Speaker_00
We have a business model that is essentially ad-supported and we force you to consume in little junks. Oh my God, they are reinventing television. It's exactly what we had with network television.

00:19:09 Speaker_00
And so then the question is, okay, so from television history, it's okay, but it's not super exciting. Where will growth come from? And I think growth will come from games.

00:19:20 Speaker_00
Activision, Blizzard, Microsoft, as so often these days, is a little ahead of the pack. But I think streaming will make use of these opportunities as well and Electronic Arts is probably one of the companies that is really attractive to acquire.

00:19:36 Speaker_03
And Felix, I think the whole streaming world is fascinating. I'm curious what the motivation is. Would a streamer want to buy them for ongoing gaming content creation or is it about the library? What are we getting if you buy Electronic Arts?

00:19:52 Speaker_03
Are you getting an engine for new games or are you getting properties?

00:19:56 Speaker_00
It's a really interesting question. In part, I think an engine for new games, that alone is attractive. The most successful gaming franchises, they have long lives, surprisingly long lives.

00:20:08 Speaker_00
And so even buying that revenue stream, unless it's priced really dearly, is attractive. But then also, maybe most important, the dividing lines between different types of content get really blurry.

00:20:22 Speaker_00
So you take gaming content and maybe you make a documentary out of it, or maybe you have a short episode where some character from a game world can appear.

00:20:33 Speaker_00
So a little bit like the blurring lines between sports and more traditional entertainment, I think the same is happening to the gaming and the more traditional video world. That makes this really attractive.

00:20:47 Speaker_03
The thing I like about this prediction is not just about gaming, I think it's a broader story about using content and distributing content as opposed to creating content.

00:20:57 Speaker_03
So I've been struck by these decisions about letting suits get onto Netflix and how powerful that has been. Historically, content creation has been the dominant way to go in the last 15 years.

00:21:10 Speaker_03
And I actually think people will scale that back and want to just monetize existing content. And I think gaming is one version of that. So I think the interesting thing about gaming is the properties and then how you monetize it.

00:21:23 Speaker_03
So in my sense of the world, we're going to end up saying, well, how can I just take existing content and monetize that as opposed to constantly asking the question of how do you produce new content?

00:21:34 Speaker_00
Interesting.

00:21:35 Speaker_00
Do you think these limits that we now see with the overviews of the Marvel universe, innumerable stories, some related to one hero or another, and maybe the aficionados can follow the storylines, but probably 90% of humanity couldn't care less.

00:21:55 Speaker_00
Do you think as we push to reuse content or as we push to reuse popular characters, that in the end the audience just gets tired of it. Your ability to do wonderful content based on the same ingredients, it's just limited.

00:22:12 Speaker_03
I think that's got to be true, but I think it's also true that new content is just not proving economically rewarding enough because of the fragmentation. It's this tension. You're absolutely right.

00:22:26 Speaker_03
Audiences want that, but it's not clear that it's going to pay off. And we just happened to live through a time when we were willing to do this remarkable level of content creation for relatively small audiences. And that just may be done.

00:22:38 Speaker_00
Yeah. Mildly depressing.

00:22:39 Speaker_03
Because I need more British cop shows.

00:22:41 Speaker_00
Yes, that's right.

00:22:44 Speaker_03
Your next prediction, Mihir. I don't know if it's a prediction or just maybe the most interesting story to watch in 2024 is going to be Argentina. It is absolutely fascinating what is going on there.

00:22:59 Speaker_03
As you I'm sure know, Felix Javier Millet has been elected. It's the new president of Argentina. To say he's a remarkable character is just a complete understatement.

00:23:08 Speaker_02
Yes.

00:23:09 Speaker_03
But the reason it's important for the world is Argentina is a serious economy. If what he is trying to do in Argentina has any kind of an effect that is salutary, I think it becomes a model for a lot of people.

00:23:25 Speaker_03
And so what he's talking about is just remarkable regulatory reform and stripping of crazy amounts of regulations. He's talked about dollarization. of an economy that's a pretty large economy.

00:23:36 Speaker_03
And so if he's even moderately successful, it's going to become a playbook of sorts. Alternatively, if he isn't,

00:23:46 Speaker_03
I think it's going to be a real tough moment for the economic populism and one version of the economic populism that has taken hold around the world.

00:23:56 Speaker_03
So I think he's going to end up doing pretty radical things, not as radical as he's promised because he'll get slowed down in various ways.

00:24:03 Speaker_03
But it is going to be a really remarkable story to watch either because I think he's going to fail significantly or he's going to have some positive benefits.

00:24:14 Speaker_00
I do not know how optimistic I should be. And not only because it's Argentina, the track record isn't exactly encouraging, but also because he lacks legislative support. So for many of the reforms that he talks about, he would need that.

00:24:32 Speaker_00
Dollarization is so difficult to even think about for an economy whose business cycles are so fundamentally different from the U.S. business cycles.

00:24:44 Speaker_00
And we spoke about it in the context of Greece, when we thought, oh, you know, if Greece had its own currency, what would the economy look like, would there be a degree of flexibility that would actually help manage the crisis.

00:24:59 Speaker_00
Now that the worst is over for Greece and Greece is now growing very quickly, it's one of the fastest growing economies in Europe, it makes you rethink that kind of discipline that comes from

00:25:12 Speaker_00
being less on your own, more constrained, maybe after all that's not such a bad thing. So the dollarization alone is completely, completely fascinating to think about.

00:25:24 Speaker_00
Also, just symbolically speaking, a clear break with the Peronist past is probably something that is really valuable for Argentina.

00:25:36 Speaker_03
Yeah. And I think the other piece of it, which is just the level of frustration in Argentina is so great because the economic performance has been so difficult.

00:25:47 Speaker_03
If you look back at a longer chart of GDP per capita in Argentina over the last hundred years, it's really quite striking and depressing. Argentina was a very wealthy country a hundred years ago.

00:26:00 Speaker_03
And it is a story that is really one of the great economic misfortunes of the last hundred years. And so then people are so desperate for such radical changes. And one has to appreciate that desperation.

00:26:15 Speaker_03
It's easy to say, oh my God, it's all kooky and crazy, but it's not because it's founded in deep, deep desperation. Seeing how that plays out, I think is going to be really fascinating.

00:26:25 Speaker_03
I'm not sure whether to be optimistic or not, Felix, but it is going to be something where We learn about some big ideas that matter for economies like Turkey, like developing countries all around the world in Africa and Latin America.

00:26:41 Speaker_03
And I think those lessons will get learned in the next 12 to 24 months.

00:26:44 Speaker_00
Yeah. Oh, my God.

00:26:45 Speaker_03
I so wish your optimism would play out. I, of course, hope for the best for Argentina, but I just think as an object lesson, it's going to turn out to be, let's put it this way, lots of dissertations are going to get written.

00:26:57 Speaker_03
Yes, one measure of output. Yeah, exactly. One measure of output is going to be going up.

00:27:12 Speaker_00
Alright Felix, what do you got? So my prediction is in 2024 we will increasingly come to see plastic as the new asbestos. I think we have taken plastic just completely for granted. At any day in your life.

00:27:31 Speaker_00
how many interactions you have with plastic products, and often you don't really think about it, and often it is just incredibly useful. It's just about perfect. And now, like in asbestos, two things have happened.

00:27:51 Speaker_00
It's a product that is related to oil and we're trying to make the economy less carbon intensive. So I think that's one headwind. That alone would probably not be that effective. But I think the conversations about microplastics

00:28:09 Speaker_00
In every human body, essentially, you find lots and lots of plastic because we're inhaling it, it's in the water, it's in the air, it's everywhere.

00:28:17 Speaker_00
I think that is essentially how asbestos came to be seen what it is to be seen, because it had this dramatic impact. on human health.

00:28:28 Speaker_00
And I think we're well on our way towards seeing plastic as something that, yes, absolutely crucial for so, so many things we do, but also deeply, deeply problematic in a way that I don't think historically we've really appreciated.

00:28:44 Speaker_03
Wow, that's a big one, Felix. I got to think about this one. The asbestos story, I don't know enough about it as I probably should, but you're right.

00:28:53 Speaker_03
It was kind of sudden and it was related to concerns about its safety that I think that really made it in some sense go away in a fast way and in a very public way.

00:29:06 Speaker_03
With plastics, there are definitely concerns and they've been percolating up for a little bit of while.

00:29:11 Speaker_03
I wonder if it requires a larger moment, something to happen where people say, oh my God, because the convenience issues and the cost efficiency is so amazing with plastics that it has to be some kind of a seismic moment where I think people wake up.

00:29:31 Speaker_03
How do we think about the replacements and the substitutes? Or do we think about just reductions in use?

00:29:36 Speaker_00
All of the above. Think about plastic straws. The image, what happens with plastic straws when they're in the water and they get eaten by fish. I think those images, I think, were absolutely crucial in convincing us.

00:29:55 Speaker_00
So first, the plastic straws get replaced by paper-based products. And then they're just awful, basically. The touch is just not right.

00:30:06 Speaker_00
And then all of a sudden we think, really, is it absolutely necessary to drink out of a glass with the help of a straw? Maybe not. Maybe we can do away with the straw altogether. And I think those are the two kinds of responses that you will see.

00:30:21 Speaker_00
So you see it in plastic bags now. We just substitute paper. And then you start thinking about, well, actually, Could I bring bags to the supermarket instead of picking up paper bags every time? Of course plastic, because it's everywhere.

00:30:38 Speaker_00
It's used across such a broad range of products. Think automobiles, basically anything you can think of, there's plastic involved. I think these substitution patterns will look very different.

00:30:51 Speaker_00
But the story that will drive it is this realization, how much we breathe it in, how much we eat it. It reminds me a little bit when you look at organic products. We made the most progress, I think, in organic food.

00:31:10 Speaker_00
And it's always this idea that, oh, it's something that's in me, or it's something that I'm feeding my kids, that then becomes a very powerful motivator.

00:31:19 Speaker_00
And I think with plastics, probably similar, where you're thinking, the moment it touches your body, the moment you ingest it in some fashion, I think that's sort of a game changer in the sense that we care about it,

00:31:33 Speaker_00
in a way that we haven't really cared about it historically. But I agree with you, the substitution process is much, much more difficult than with asbestos, where we used to think of asbestos as just this amazing building material.

00:31:48 Speaker_00
We forget now how even very late companies made acquisitions of other companies that had asbestos legacies, not really thinking about the long-term consequences.

00:32:00 Speaker_00
Plastic is sort of at this moment where you can think about it and, you know, it's completely innocent and who cares? And I think you will come to regret it.

00:32:09 Speaker_03
That's great.

00:32:10 Speaker_00
Mihir, what do you have?

00:32:12 Speaker_03
Well, the last decade has seen this remarkable rise in quantitative trading and algorithmic trading. And I think in part, we're ready for something to shake that world.

00:32:25 Speaker_03
And I'm not sure what will shake that world, but I think it is likely to be some kind of a scandal and maybe regulatory crackdown. So these quant funds are remarkable players in financial markets today and their growth has kind of

00:32:42 Speaker_03
happened without us really knowing fully how to think about them.

00:32:46 Speaker_03
I'm not speaking entirely of just index funds, things like Vanguard and BlackRock, which have already seen a little bit of pushback, but I'm really talking about quantitative systematic trading funds.

00:32:58 Speaker_03
And we referenced this, Felix, in our last episode when we talked a little bit about how market movements appear to be out of line with things that are going on in the economy. And

00:33:09 Speaker_03
I think there's conceivably a regulatory crackdown and maybe a market manipulation scandal that will plague that world and will cause people to ask questions that have not been asked during the last 15 years.

00:33:25 Speaker_03
So in particular, what kind of asset prices are contributing to their returns? Which kinds of assets? And what are they doing? And what are those strategies? And so once regulators get interested in those questions,

00:33:39 Speaker_03
I think that world will have to kind of become much more public and much more mainstream. And in that process, I think we may learn things that people may feel not so great about.

00:33:50 Speaker_03
And so I'm not exactly sure how that shakes out, Felix, as you can tell from my vagueness, but my sense is that that world is ripe for some combination of a scandal or regulatory crackdown that changes the rules of the game in interesting ways.

00:34:08 Speaker_00
Is your sense that there are particular asset classes that are problematic or is it that trading practices and the extent to which they exploit momentum that then leads us to asking questions about where the momentum comes from in the first place?

00:34:27 Speaker_00
It's that piece, Felix. Exactly. Thank you. You said it better than I did.

00:34:30 Speaker_03
which is I think there's gonna be a sense of what is going on with momentum in markets and who's driving that momentum. And then are we seeing more violence in markets than we need to?

00:34:41 Speaker_03
Because in fact, these are self-reinforcing, self-feeding phenomenon and the detachment from fundamentals in certain asset classes. And of course, we've seen this before.

00:34:52 Speaker_03
The old version of this was market manipulation scandals in the silver market where you had like the Hunt brothers doing things. I'm just wondering if there's something like that that is going to end up playing out.

00:35:04 Speaker_03
I know that's a very vague idea, but I have a feeling that something like that can happen, especially given what's been going on in the last 12 or 18 months.

00:35:13 Speaker_00
Yeah. One reason why I find this prediction so interesting is because it will also lead to a conversation about how the trading is related to real economic activity.

00:35:25 Speaker_03
And I think that's the triggering thing, which I think we end up asking those questions, which we should be asking. And frankly, that world has grown up without asking those questions. I think we're ripe to see something there, Felix.

00:35:35 Speaker_00
Yeah. Wow. Okay. Excellent. It'll be an exciting 2024. Exactly.

00:35:40 Speaker_03
Exactly.

00:35:40 Speaker_00
And what do you have, Felix? I have a slightly gloomier one. I think the quality of search results, so Google search or Bing or any one of these will deteriorate rapidly. Even more so, you mean? Even more so than today.

00:35:57 Speaker_00
So today, frankly, it's basically the business model that pollutes the search results. 15,000 sponsored links that are vaguely unrelated to what you're really looking for, so that you have to now go on page five to really figure out what you want.

00:36:16 Speaker_00
So I think that's not going to disappear, but I think it will be incredibly difficult for Google and everyone in the search business to distinguish between AI generated content.

00:36:29 Speaker_00
that then includes every hallucination of AI that you can imagine, and the content that is verified high quality. This is such a lucrative market. Search is just so incredibly profitable.

00:36:45 Speaker_00
Think about all the industries that sprang up around search engine optimization.

00:36:50 Speaker_00
That is going to have an exact equivalent in how can I produce AI content that is really hard to detect for Google, that is really hard to detect for Microsoft, because that's like search engine optimization, you end up

00:37:07 Speaker_00
in the results, and that is of course where the big money is. As a result, if you want to see a silver lining in this, I think we will rely to a much greater extent on trusted brands.

00:37:25 Speaker_00
I would not be surprised if, say, the use of Wikipedia would explode. because it's a source that is less polluted. Normally, people are skeptical about what AI means for search, but they're thinking that AI chatbots will replace search.

00:37:44 Speaker_00
But I think that's actually not quite right. I think what will happen is search itself will be such a miserable experience as a result of AI that you'll go directly to the New York Times or you'll go directly to the Wall Street Journal.

00:37:59 Speaker_03
Wow, that's a radical prediction. I love this prediction. Let me push back a little, which is, it seems like it becomes a cat and mouse game between Google and all these people trying to trick Google. And if that's the cat and mouse game,

00:38:15 Speaker_03
Isn't Google gonna win? Aren't they gonna be able to like figure it out? If it is this cat and mouse game, do we really think that these more bit players can outfox Google given how valuable the search business is to them?

00:38:30 Speaker_00
Why can't they always stay one step ahead? So think about, practically speaking, how that might happen. So you have a piece of content that lives on the web already.

00:38:41 Speaker_00
And now we're thinking, how would we distinguish something that has been generated by AI from something that wasn't AI generated in the first place, which might just be wrong information. You could exclude particular sources,

00:38:58 Speaker_00
So you could say anything on Reddit. Reddit is probably going to be completely overwhelmed by bots. And as a result, anything on Reddit, I don't trust. That makes search worse, because guess what?

00:39:12 Speaker_00
There is a lot of really fabulous content on Reddit that should be included. You could do things like if the grammar is not quite right. We know that all of these chatbots are really beautiful at grammar.

00:39:25 Speaker_00
I now get email from people who English is not their first language and they're writing me these really beautiful emails, which obviously come from some chatbot. So then what will people do? They will introduce mistakes in their AI generated content.

00:39:41 Speaker_00
So in search engine optimization, it's just been incredibly, incredibly hard to beat the optimizers at their game. That's interesting. No real reason to believe that that should be different in AI.

00:39:57 Speaker_03
Right. And so it is just this remarkable cheapening of the technology to create this stuff.

00:40:03 Speaker_00
Yes.

00:40:03 Speaker_03
It's a persistent problem, but now that the costs of producing that bad content have gone so low that the battle shifts in some tectonic way.

00:40:13 Speaker_00
That's interesting. 97% of the web will be bot generated and so then search really becomes a search for the needle in the haystack.

00:40:27 Speaker_03
Okay, that's not something to look forward to.

00:40:28 Speaker_00
Yes. Hold on to your New York Times subscription. I think it's one piece of advice. There you go. I'm fresh out of predictions for 2024. How about you, Mir? Do you have one or two for us?

00:40:41 Speaker_03
I just have three little weird product kinds of recommendations or thoughts about what will be important. So I'll give you all three super fast. One, I think energy drinks are gonna really struggle.

00:40:54 Speaker_03
There have been now more and more health concerns that have been raised. So for example, Panera has run into this with their lemonade and they've stuck by it, interestingly. And I think they get replaced by prebiotic drinks.

00:41:07 Speaker_03
So this is my beverage market prediction, which is energy drinks finally have their comeuppance and prebiotic drinks, which are basically drinks with fiber embedded in them, kind of take them.

00:41:19 Speaker_03
Number two, I think the Apple Vision Pro is going to be amazing. It's going to be more amazing than people realize. Forecasts are for something like maybe 100,000 to maybe 400,000 units in the first year.

00:41:35 Speaker_03
And I have a feeling it's going to just be like more than a million. Now, that doesn't really move the needle for Apple in any meaningful way. And I don't think it becomes a product. like the iPhone or anything else.

00:41:46 Speaker_03
If you've seen the video, the way they integrate the real world and that world, it's really something else. And so I think that's going to be better than you think.

00:41:55 Speaker_00
Wow.

00:41:55 Speaker_03
I see someone spend thousands of dollars on a new headset. You're right. It's going to be around a $4,000 price tag. And I think that market is larger than we understand it to be. And it's going to be far better than Meta's efforts.

00:42:11 Speaker_03
It's going to be far better than a lot of people's efforts. So that'll be interesting. And the final one is Percy Jackson, new Disney content is going to go gangbusters and is going to lead to like a whole revival of Disney.

00:42:26 Speaker_03
So yeah, there you go, three crazy product predictions.

00:42:29 Speaker_00
Can I ask you about the Apple prediction? I can completely see why it's so novel, it's so good, lots of people will want one. The bigger question I think is, do you see people use it daily? You know how that novel thing that is amazing and then Really?

00:42:51 Speaker_00
You want to wear something like that every day for long hours?

00:42:55 Speaker_03
Well, so how about this? Obviously gaming is a big part of this, which I'm not going to talk about, but let's just think about consuming content on an iPad. So I will watch a TV show or a movie on an iPad.

00:43:07 Speaker_03
I'm not sure I'm going to be doing that in a couple years. I actually think I could be using the Apple Vision Pro to be consuming content in that way. And it'll be a better visual experience. Yeah.

00:43:18 Speaker_03
And I'll have integrated with the real world in a kind of interesting way. And I think it replaces that form of content consumption. And that is a large market.

00:43:29 Speaker_02
Yeah.

00:43:29 Speaker_03
By the way, I'm not making predictions about that. I just think in the short run, it's going to be a banging product and it's going to be way better than the competition. And then ultimately, it's about

00:43:40 Speaker_03
Am I willing to switch to that way to consume content? Pure like sit back kind of consume content, not lean forward content consumption. And I think it can be. So that I think is going to be fun to watch.

00:43:50 Speaker_00
It's interesting. I do think comfort will decide its fate. Exactly. You know, I've been wearing glasses forever and my glasses are not uncomfortable in any real way, but do I like moments when I don't have to wear my glasses? Yes, absolutely.

00:44:07 Speaker_00
So somehow the comfort level needs to be quite amazing for it to make it into an everyday product, in particular the moment when you want to relax and watch a show.

00:44:18 Speaker_03
Well, that's the interesting part. I don't think there's a world where we're all walking around with this stuff on, at least not now. But that can happen, maybe.

00:44:25 Speaker_03
I think it's a little bit more about, do I consume a little bit of content for an hour or two at night? And is that an interesting way to do it? And I think it may well be. It's been a while since I've done the Apple fanboy thing. So there you go.

00:44:48 Speaker_03
Okay, Felix, recommendations. What do you have for us this week?

00:44:52 Speaker_00
One of the trends I paid a little bit of attention to is the quality of data visualization. And it's been around as a topic for quite some time now. But I have to say, wherever you look, mainstream media, I follow a subreddit called Data is Beautiful.

00:45:14 Speaker_00
People have just become so good at telling sometimes complex stories with the help of beautiful and simple data visualization. I wanted to recommend one particular graph that I really loved. It appeared in The Economist.

00:45:31 Speaker_00
And it asks the question, how rich are countries really? If we had a ranking of the richest country, what would it look like? And what the graph does is it takes three steps. It starts just with GDP per capita.

00:45:46 Speaker_00
And then it asks, well, but actually prices of products and services are not the same. What if we adjust for Switzerland being super super expensive and other economies less so?

00:45:57 Speaker_00
And then the last step is an adjustment for how many hours do people actually work to produce that kind of income. And it's one of these things where just…

00:46:09 Speaker_00
seeing these very simple and obvious ideas, how big a difference they make, is just absolutely fabulous. So the US, for instance, starts out ranked seventh on the list of GDP per capita. It slips by a rank once you take prices into account.

00:46:27 Speaker_00
And then it ends up being 11th once you take into account that Americans work long hours. One of the most dramatic changes is if you look at Belgium, no one thinks of Belgium as a superstar economy, so they start out 18th.

00:46:46 Speaker_00
on GDP per capita, they're 15th once you take prices into account. And then once you take into account how much they work, they end up being the sixth richest economy.

00:46:58 Speaker_00
So it's all captured in this really simple, elegant graph that connects the different steps. People who do data visualization, If you have a chance to look at it on Visual Capitalist, on Reddit, on any of these forums, I think it's such a joy.

00:47:17 Speaker_00
You learn so much and it's often just incredibly beautiful to look at at the same time.

00:47:24 Speaker_03
That is a great recommendation. I agree. Data visualization is such a wonderful field. I remember being obsessed with Tufti way back when. The FT actually has gotten really good at this.

00:47:33 Speaker_03
And one of the things they've done, which I really love, is the regional differences within countries. So if you look within the U.S. and you look at the poorest parts of those countries, they look a lot like developing countries all around the world.

00:47:48 Speaker_03
Looking under the hood is so powerful, just like what your example does, which is really fantastic. That sounds great. What do you have for us, Mihir? So I have just started this book, but I have just already fallen in love with it.

00:48:00 Speaker_03
It is a book by Carlo Rovelli, who is a physicist who writes books for the lay audience about physics.

00:48:07 Speaker_03
But his latest book is about Anaximander, which is a mouthful and somebody I had never heard of before, who is a Greek thinker and really Rovelli argues kind of the original scientific thinker.

00:48:22 Speaker_03
And so Anaximander turns out to be this incredible character who arguably is the father of the scientific method, is somebody who thought about evolution quite deeply early, who thought about laws. And I had never heard his name before.

00:48:41 Speaker_03
I haven't either, yeah. And Rovelli writes so beautifully. And in a world that sometimes is lacking reason, it is wonderful to read about somebody who thinks so seriously about the origins of science.

00:48:54 Speaker_03
So it's Carlo Rovelli, and the book is called Anoxymander and the Birth of Science. And it's really fantastic.

00:49:01 Speaker_00
Fabulous. So this is it for tonight and this is it for 2023. We wanted to thank all our listeners for listening in, for thinking with us about the world and the changes in business, the changes in culture.

00:49:19 Speaker_00
And then, of course, a very, very big thank you to our audio engineer, Peter Linain, who's responsible for the wonderful sounds that you get to hear on the After Hours podcast. This is it. After Hours from the TED Audio Collective.